Analysis CLL Proteomics - Final data Dimension Reduction
Sophie Rabe
Analysis date: 2020-01-10
Setup
Load libraries
library(plyr)
library(gtools)
library(openxlsx)
library(pheatmap)
library(reshape2)
library(progress)
library(Matrix)
library(Hmisc)
library(lemon)
library(ggpubr)
library(effsize)
library(ggbeeswarm)
library(ggfortify)
library(ggpmisc)
library(ggrepel)
library(readxl)
library(DESeq2)
library(TOSTER)
library(tidyverse)
library(vsn)
library(fdrtool)
library(limma)
library(apeglm)
library(IHW)
library(Rtsne)
library(biomartr)
library(biomaRt)
library(MultiAssayExperiment)
library(PMA)
library(gplots)
library(RColorBrewer)
library(grid)
library(ConsensusClusterPlus)
library(survival)
library(survminer)
library(cowplot)
library(viridis)Load data
source("/Volumes/sd17b003/Sophie/Analysis/Screen_analysis/Figure_layouts.R")
load("/Volumes/sd17b003/Sophie/Analysis/CLL_Proteomics/CLL_Proteomics_final/Proteomics_Git/Robjects/CLL_Proteomics_Setup.RData")
load("/Volumes/sd17b003/Sophie/Analysis/CLL_Proteomics/CLL_Proteomics_final/Proteomics_Git/Robjects/CLL_Proteomics_ConsensusClustering.RData")
load("/Volumes/sd17b003/Sophie/Analysis/CLL_Proteomics/R_objects/GOterm_BCR.RData")
load("/Volumes/sd17b003/Sophie/Analysis/CLL_Proteomics/R_objects/GOterm_Spliceosome.RData")colData(multiomics_MAE)$PG <- as.factor(CCP_group5[rownames(colData(multiomics_MAE))])
colData(multiomics_MAE)$CCP6_RNA <- as.factor(CCP_group6_RNA[rownames(colData(multiomics_MAE))])Analysis
Dimension reduction plots
tsne (Barnes-Hut-SNE)
t-SNE colored by genetic alterations and groups
set.seed(10)
rtsne_out <- Rtsne( t(assay(multiomics_MAE[prot_few_nas , ,"proteomics"])), perplexity = 10 )
rtsne_out_df <- rtsne_out$Y %>% as.data.frame() %>% as_tibble()
rtsne_out_df$pat_ID <- colnames(assay(multiomics_MAE[prot_few_nas , ,"proteomics"]))
rtsne_out_df <- left_join(rtsne_out_df,
wideFormat(multiomics_MAE[, ,c("SNPs","chrom_abber", "health_record_bin") ], colDataCols = c("gender", "treatment_status", "doehner_groups", "PG")) %>% as_tibble(),
by=c("pat_ID"="primary") )
rtsne_out_df <- mutate_at(rtsne_out_df, colnames(rtsne_out_df %>% dplyr::select(SNPs_ATM:health_record_bin_treated)), as.logical)
rtsne_out_df <- rtsne_out_df %>% replace(is.na(.), "unknown")
message("t-SNE colored by Döhner groups")
ggplot(rtsne_out_df, aes(V1, V2)) +
geom_point(aes(color=doehner_groups), size=3) +
pp_sra +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
message("t-SNE colored by consensus cluster groups")
tsne_CCP_P_plot <- ggplot(rtsne_out_df, aes(V1, V2)) +
geom_point(aes(fill=PG), color="grey", shape=21) +
scale_fill_manual(values = colors_CCP)+
pp_sra +
guides(color=guide_legend(title="PG"))
tsne_CCP_P_plot+ theme(aspect.ratio=1, plot.title = element_text(size = 30))
message("t-SNE colored by combination of IGHV status and trisomy12")
tsne_IGHVtris_plot <- ggplot(rtsne_out_df %>% filter(health_record_bin_IGHV_mutated!="unknown", chrom_abber_trisomy12!="unknown"), aes(V1, V2)) +
geom_point(aes(fill=interaction(health_record_bin_IGHV_mutated, chrom_abber_trisomy12)), color="grey", shape=21) +
scale_fill_manual(values = colors_CCP[c(1,3,4,2)] )+
pp_sra_noguides
tsne_IGHVtris_plot + theme(aspect.ratio=1, plot.title = element_text(size = 30)) #+
message("t-SNE colored by IGHV status")
ggplot(rtsne_out_df, aes(V1, V2)) +
geom_point(aes_string(color="health_record_bin_IGHV_mutated"), size=3) +
scale_color_manual(values = c( "#0571b0", "#ca0020", "grey")) +
pp_sra +
theme(aspect.ratio=1, plot.title = element_text(size = 30))+
guides(color=guide_legend(title="IGHV_mutated"))
message("t-SNE colored by trisomy12")
ggplot(rtsne_out_df, aes(V1, V2)) +
geom_point(aes_string(color="chrom_abber_trisomy12"), size=3) +
scale_color_manual(values = c( "#0571b0", "#ca0020", "grey")) +
pp_sra +
theme(aspect.ratio=1, plot.title = element_text(size = 30))+
guides(color=guide_legend(title="trisomy12"))
sapply(colnames(rtsne_out_df)[c(4,5,8:29)], function(var){
print(ggplot(rtsne_out_df, aes(V1, V2)) +
geom_point(aes_string(color=var), size=3) +
scale_color_manual(values = c( "#0571b0", "#ca0020", "grey")) +
ggtitle(gsub("SNPs_", "", gsub("chrom_abber_", "", gsub( "health_record_bin_", "", var) ))) +
pp_sra_noguides +
theme(aspect.ratio=1, plot.title = element_text(size = 30)) )
})
## gender treatment_status SNPs_ATM SNPs_BIRC3 SNPs_EGR2 SNPs_NOTCH1
## data List,29 List,29 List,29 List,29 List,29 List,29
## layers List,1 List,1 List,1 List,1 List,1 List,1
## scales ? ? ? ? ? ?
## mapping List,2 List,2 List,2 List,2 List,2 List,2
## theme List,66 List,66 List,66 List,66 List,66 List,66
## coordinates ? ? ? ? ? ?
## facet ? ? ? ? ? ?
## plot_env ? ? ? ? ? ?
## labels List,4 List,4 List,4 List,4 List,4 List,4
## guides List,4 List,4 List,4 List,4 List,4 List,4
## SNPs_POT1 SNPs_SF3B1 SNPs_TP53 SNPs_XPO1 chrom_abber_del11q
## data List,29 List,29 List,29 List,29 List,29
## layers List,1 List,1 List,1 List,1 List,1
## scales ? ? ? ? ?
## mapping List,2 List,2 List,2 List,2 List,2
## theme List,66 List,66 List,66 List,66 List,66
## coordinates ? ? ? ? ?
## facet ? ? ? ? ?
## plot_env ? ? ? ? ?
## labels List,4 List,4 List,4 List,4 List,4
## guides List,4 List,4 List,4 List,4 List,4
## chrom_abber_del13q14 chrom_abber_del17p13 chrom_abber_del5_IgH
## data List,29 List,29 List,29
## layers List,1 List,1 List,1
## scales ? ? ?
## mapping List,2 List,2 List,2
## theme List,66 List,66 List,66
## coordinates ? ? ?
## facet ? ? ?
## plot_env ? ? ?
## labels List,4 List,4 List,4
## guides List,4 List,4 List,4
## chrom_abber_delIgH_break chrom_abber_gain14q32 chrom_abber_gain8q24
## data List,29 List,29 List,29
## layers List,1 List,1 List,1
## scales ? ? ?
## mapping List,2 List,2 List,2
## theme List,66 List,66 List,66
## coordinates ? ? ?
## facet ? ? ?
## plot_env ? ? ?
## labels List,4 List,4 List,4
## guides List,4 List,4 List,4
## chrom_abber_trisomy12 health_record_bin_elderly_at_diagnosis
## data List,29 List,29
## layers List,1 List,1
## scales ? ?
## mapping List,2 List,2
## theme List,66 List,66
## coordinates ? ?
## facet ? ?
## plot_env ? ?
## labels List,4 List,4
## guides List,4 List,4
## health_record_bin_elderly_patient health_record_bin_gender_binary
## data List,29 List,29
## layers List,1 List,1
## scales ? ?
## mapping List,2 List,2
## theme List,66 List,66
## coordinates ? ?
## facet ? ?
## plot_env ? ?
## labels List,4 List,4
## guides List,4 List,4
## health_record_bin_IGHV_mutated
## data List,29
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4
## health_record_bin_komplex_abberant_karyotype
## data List,29
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4
## health_record_bin_treated
## data List,29
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4message("There is are trisomy12 negative patients which clusters with all of the other trisomy12 patients. Does they have a subclonal mutations?")
metadata(multiomics_MAE)$fish_df_clonsizes["trisomy12" ,] %>%
dplyr::select(rtsne_out_df %>% filter(chrom_abber_trisomy12==FALSE) %>% arrange(desc(V1)) %>% slice(1:2) %>% .$pat_ID)BCR signaling
Different kinds of plots
BCR_genes_mean <- assay(multiomics_MAE[BCR_genes$symbol, ,"proteomics"]) %>% colMeans(na.rm = TRUE) %>% enframe()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'BCR_CCP_P_plot <- left_join(BCR_genes_mean,
enframe(multiomics_MAE$PG, value = "PG"),
by=c("name")) %>%
filter(!is.na(PG)) %>%
ggplot(aes( PG, value ))+
geom_boxplot(aes(fill=PG)) + geom_beeswarm() +
ggtitle("BCR protein abundance") +
scale_fill_manual(values=colors_CCP) +
pp_sra +
ylab("Mean abundance of BCR proteins")+
#theme(legend.position = "bottom") +
stat_compare_means(label = "p.signif", method = "t.test",
ref.group = ".all.", label.y = 0.2, hide.ns = TRUE) +
guides(fill=guide_legend(title="PG"))
BCR_CCP_P_plot + theme(aspect.ratio=1) +
stat_compare_means(method = "anova", label.y = 0.22, hjust=0)
Heatmaps
Selected BCR proteins
# Smaller heatmap with only selected proteins
#sel_BCR <- c("ZAP70", "IGHM", "CD79A", "CD79B", "LYN", "SYK", "BLNK", "PLCG2", "BTK", "PTPN6", "NFATC2",
# "MAPK1", "MAP2K2", "NRAS", "MALT1", "BCL10", "PIK3CD","CD19", "VAV3",
# "AKT1", "IKBKB")
sel_BCR <- c("ZAP70", "IGHM", "CD79A", "CD79B", "SYK", "PLCG2", "BTK", "PTPN6",
"MAPK1", "PIK3CD", "AKT1", "IKBKB")
tmp_BCR <- wideFormat(multiomics_MAE[sel_BCR, ,"proteomics"], colDataCols = c("PG", "IGHV" )) %>%
as_tibble()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'tmp_BCR_mx <- tmp_BCR %>% dplyr::select(-primary, -PG, -IGHV) %>% as.matrix()
rownames(tmp_BCR_mx) <- tmp_BCR$primary
colnames(tmp_BCR_mx) <- gsub("proteomics_", "", colnames(tmp_BCR_mx))
tmp_BCR_anno <- tmp_BCR[, c("PG", "IGHV")] %>% as.data.frame()
rownames(tmp_BCR_anno) <- tmp_BCR$primary
tmp_BCR_anno$IGHV[tmp_BCR_anno$IGHV=="U"] <- "U-CLL"
tmp_BCR_anno$IGHV[tmp_BCR_anno$IGHV=="M"] <- "M-CLL"
ann_colors = list(
PG=c("1"= colors_CCP[1], "2"= colors_CCP[2], "3"= colors_CCP[3], "4"= colors_CCP[4], "5"= colors_CCP[5], "6"= colors_CCP[6] ),
IGHV=c("U-CLL"= "#0571b0", "M-CLL"= "#ca0020"))
tmp_BCR_anno$PG <- factor(tmp_BCR_anno$PG, levels = c(1:3,5,6,4))
breaks= seq(min(tmp_BCR_mx), max(tmp_BCR_mx), 0.1)^2
breaks= sort(c(-breaks, breaks))
breaks <- breaks[! (breaks < min(tmp_BCR_mx) | breaks > max(tmp_BCR_mx) )]
pat_order_hclust <- sapply(c(1:3,5,6,4), function(P){
hc <- hclust(dist(tmp_BCR_mx[rownames(tmp_BCR_anno[tmp_BCR_anno$PG==P,]), ] ))
hc$labels[hc$order]
}) %>% unlist
PG_BCR_proteins_pheatmap <- pheatmap(t(tmp_BCR_mx[pat_order_hclust, ]),
annotation_col = tmp_BCR_anno, annotation_colors = ann_colors, scale = "row", cluster_cols = FALSE,
color = inferno(length(breaks)), border_color = NA,
gaps_col = (which(!tmp_BCR_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$PG %>% duplicated())-1)[-1],
breaks = breaks , cutree_rows = 4, show_colnames = F, treeheight_row = 0, fontsize_row = 5)
GO terms B-cell activation or BCR signaling combined
tmp_BCR_GO <- wideFormat(multiomics_MAE[unique(c(GO_BC_activation$hgnc_symbol, GO_BCR$hgnc_symbol)), ,"proteomics"], colDataCols = c("PG", "IGHV" )) %>%
as_tibble()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'tmp_BCR_GO_mx <- tmp_BCR_GO %>% dplyr::select(-primary, -PG, -IGHV) %>% as.matrix()
rownames(tmp_BCR_GO_mx) <- tmp_BCR_GO$primary
colnames(tmp_BCR_GO_mx) <- gsub("proteomics_", "", colnames(tmp_BCR_GO_mx))
tmp_BCR_GO_anno <- tmp_BCR_GO[, c("PG", "IGHV")] %>% as.data.frame()
rownames(tmp_BCR_GO_anno) <- tmp_BCR_GO$primary
tmp_BCR_GO_anno$PG <- factor(tmp_BCR_GO_anno$PG, levels = c(1:3,5,6,4))
ann_colors = list(
PG=c("1"= colors_CCP[1], "2"= colors_CCP[2], "3"= colors_CCP[3], "4"= colors_CCP[4], "5"= colors_CCP[5], "6"= colors_CCP[6] ))
breaks= seq(min(tmp_BCR_GO_mx), max(tmp_BCR_GO_mx), 0.1)^2
breaks= sort(c(-breaks, breaks))
breaks <- breaks[! (breaks < min(tmp_BCR_GO_mx) | breaks > max(tmp_BCR_GO_mx) )]
message("Heatmap of abundance BCR signaling proteins ordered according to PG")## Heatmap of abundance BCR signaling proteins ordered according to PGPG_BCR_pheatmap <- pheatmap(t(tmp_BCR_GO_mx[(tmp_BCR_GO_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$rowname), ]),
annotation_col = tmp_BCR_GO_anno, annotation_colors = ann_colors, scale = "row", cluster_cols = FALSE,
color = inferno(length(breaks)), border_color = NA,
gaps_col = (which(!tmp_BCR_GO_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$PG %>% duplicated())-1)[-1],
breaks = breaks , cutree_rows = 3, show_colnames = F, treeheight_row = 0, fontsize_row = 5)
Spliceosome
KEGG
splice_genes_mean <- assay(multiomics_MAE[splice_genes$symbol, ,"proteomics"]) %>% colMeans(na.rm = TRUE) %>% enframe()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'splicing_CCP_P_plot <- left_join(splice_genes_mean,
enframe(multiomics_MAE$PG, value = "PG"),
by=c("name")) %>%
filter(!is.na(PG)) %>%
ggplot(aes( PG, value ))+
geom_boxplot(aes(fill=PG)) + geom_beeswarm() +
ggtitle("Spliceosome protein abundance") +
pp_sra +
ylab("Mean abundance of spliceosome proteins")+
stat_compare_means(label = "p.signif", method = "t.test",
ref.group = ".all.", label.y = 0.2, hide.ns = TRUE)+
guides(fill=guide_legend(title="PG")) +
scale_fill_manual(values = colors_CCP)
splicing_CCP_P_plot + theme(aspect.ratio=1) +
stat_compare_means(method = "anova", label.y = 0.22, hjust=0)
GO term Spliceosomal complex
tmp_Splice_GO <- wideFormat(multiomics_MAE[GO_SpliceosomalComplex$hgnc_symbol, ,"proteomics"], colDataCols = c("PG", "IGHV" )) %>%
as_tibble()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'tmp_Splice_GO_mx <- tmp_Splice_GO %>% dplyr::select(-primary, -PG, -IGHV) %>% as.matrix()
rownames(tmp_Splice_GO_mx) <- tmp_Splice_GO$primary
colnames(tmp_Splice_GO_mx) <- gsub("proteomics_", "", colnames(tmp_Splice_GO_mx))
tmp_Splice_GO_anno <- tmp_Splice_GO[, c("PG", "IGHV")] %>% as.data.frame()
rownames(tmp_Splice_GO_anno) <- tmp_Splice_GO$primary
tmp_Splice_GO_anno$PG <- factor(tmp_Splice_GO_anno$PG,levels = c(1:3,5,6,4))
ann_colors = list(
PG=c("1"= colors_CCP[1], "2"= colors_CCP[2], "3"= colors_CCP[3], "4"= colors_CCP[4], "5"= colors_CCP[5], "6"= colors_CCP[6] ))
breaks= seq(min(tmp_Splice_GO_mx), max(tmp_Splice_GO_mx), 0.1)^2
breaks= sort(c(-breaks, breaks))
breaks <- breaks[! (breaks < min(tmp_Splice_GO_mx) | breaks > max(tmp_Splice_GO_mx) )]
message("Heatmap of abundance spliceosome proteins ordered according to PG")## Heatmap of abundance spliceosome proteins ordered according to PGPG_GOSplice_pheatmap <- pheatmap(t(tmp_Splice_GO_mx[(tmp_Splice_GO_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$rowname), ]),
annotation_col = tmp_Splice_GO_anno, annotation_colors = ann_colors, scale = "row", cluster_cols = FALSE,
color = inferno(length(breaks)), border_color = NA,
gaps_col = (which(!tmp_Splice_GO_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$PG %>% duplicated())-1)[-1],
breaks = breaks , show_colnames = F , treeheight_row = 0, fontsize_row = 2)
Selected spliceosome proteins
some_spliceprots <- c("SF3B1", "SNRPA", "PRPF6", "PRPF3", "SF3A1", "SNRPD2", "SRSF4", "CDC5L", "PRPF19",
"CRNKL1", "PUF60", "PRPF8",
"SNRPB2")
tmp_Splice_GO <- wideFormat(multiomics_MAE[some_spliceprots, ,"proteomics"], colDataCols = c("PG", "IGHV" )) %>%
as_tibble()## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'tmp_Splice_GO_mx <- tmp_Splice_GO %>% dplyr::select(-primary, -PG, -IGHV) %>% as.matrix()
rownames(tmp_Splice_GO_mx) <- tmp_Splice_GO$primary
colnames(tmp_Splice_GO_mx) <- gsub("proteomics_", "", colnames(tmp_Splice_GO_mx))
tmp_Splice_GO_anno <- tmp_Splice_GO[, c("PG", "IGHV")] %>% as.data.frame()
rownames(tmp_Splice_GO_anno) <- tmp_Splice_GO$primary
tmp_Splice_GO_anno$PG <- factor(tmp_Splice_GO_anno$PG,levels = c(1:3,5,6,4))
SF3B1mut <- wideFormat(multiomics_MAE["SF3B1",,"SNPs"])$SNPs_SF3B1## harmonizing input:
## removing 1 colData rownames not in sampleMap 'primary'names(SF3B1mut) <- wideFormat(multiomics_MAE["SF3B1",,"SNPs"])$primary## harmonizing input:
## removing 1 colData rownames not in sampleMap 'primary'tmp_Splice_GO_anno$SF3B1 <- as.factor(SF3B1mut[rownames(tmp_Splice_GO_anno)])
tmp_Splice_GO_anno$IGHV[tmp_Splice_GO_anno$IGHV=="U"] <- "U-CLL"
tmp_Splice_GO_anno$IGHV[tmp_Splice_GO_anno$IGHV=="M"] <- "M-CLL"
ann_colors = list(
PG=c("1"= colors_CCP[1], "2"= colors_CCP[2], "3"= colors_CCP[3], "4"= colors_CCP[4], "5"= colors_CCP[5], "6"= colors_CCP[6] ),
SF3B1=c("1"="darkblue", "0"="gray80"),
IGHV=c("U-CLL"= "#0571b0", "M-CLL"= "#ca0020"))
pat_order_hclust_splice <- sapply(c(1:3,5,6,4), function(P){
hc <- hclust(dist(tmp_Splice_GO_mx[rownames(tmp_Splice_GO_anno[tmp_Splice_GO_anno$PG==P,]), ] ))
hc$labels[hc$order]
}) %>% unlist
PG_splice_proteins_pheatmap <- pheatmap(t(tmp_Splice_GO_mx[pat_order_hclust_splice, ]),
annotation_col = tmp_Splice_GO_anno, annotation_colors = ann_colors, scale = "row", cluster_cols = FALSE,
color = inferno(length(breaks)), border_color = NA,
gaps_col = (which(!tmp_Splice_GO_anno %>% rownames_to_column() %>% arrange(PG, IGHV) %>% .$PG %>% duplicated())-1)[-1],
breaks = breaks , show_colnames = F , treeheight_row = 0, fontsize_row = 5)
PCA
prot_pca <- prcomp(t( assay(multiomics_MAE[prot_few_nas , ,"proteomics"]) ))## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'prot_pca_x <- as_tibble(prot_pca$x[,1:10])
prot_pca_x$pat_ID <- colnames(assay(multiomics_MAE[prot_few_nas , ,"proteomics"]))## harmonizing input:
## removing 13 colData rownames not in sampleMap 'primary'prot_pca_x <- left_join(prot_pca_x,
wideFormat(multiomics_MAE[, ,c("SNPs","chrom_abber", "health_record_bin") ],
colDataCols = c("gender", "treatment_status", "PG")) %>% as_tibble() ,
by=c("pat_ID"="primary") )## ExperimentList contains data.frame or DataFrame,
## potential for errors with mixed data typesprot_pca_x <- prot_pca_x %>% replace(is.na(.), "unknown")
ggplot(prot_pca_x, aes(PC1, PC2)) +
geom_point(aes(color=health_record_bin_IGHV_mutated, shape=chrom_abber_trisomy12)) + pp_sra +
scale_shape_manual(values = c( 16, 1, 4)) +
scale_color_manual(values = c( "#0571b0","#ca0020", "grey")) + theme(aspect.ratio = 1)
ggplot(prot_pca_x, aes(PC1, PC3)) +geom_point(aes(color=health_record_bin_IGHV_mutated, shape=chrom_abber_trisomy12)) +
scale_shape_manual(values = c( 16, 1, 4)) +
scale_color_manual(values = c( "#0571b0","#ca0020", "grey"))+ pp_sra+ theme(aspect.ratio = 1)
ggplot(prot_pca_x, aes(PC2, PC3)) +geom_point(aes(color=health_record_bin_IGHV_mutated, shape=chrom_abber_trisomy12)) +
scale_shape_manual(values = c( 16, 1, 4)) +
scale_color_manual(values = c( "#0571b0", "#ca0020", "grey")) + pp_sra+ theme(aspect.ratio = 1)
ggplot(prot_pca_x, aes(PC1, PC2)) +geom_point(aes(color=gender)) + pp_sra + scale_color_manual(values = c( "#0571b0","#ca0020", "grey")) + theme(aspect.ratio = 1)
PCA_CCP_1_2 <- ggplot(prot_pca_x, aes(PC1, PC2)) +
geom_point(aes(fill=PG),shape=21, color="grey") + pp_sra +
guides(color=guide_legend(title="PG")) +
scale_fill_manual(values = colors_CCP)
PCA_CCP_1_2 +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
PCA_CCP_1_3 <- ggplot(prot_pca_x, aes(PC1, PC3)) +
geom_point(aes(fill=PG),shape=21, color="grey") + pp_sra +
guides(color=guide_legend(title="PG")) +
scale_fill_manual(values = colors_CCP)
PCA_CCP_1_3 +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
PCA_CCP_2_3 <- ggplot(prot_pca_x, aes(PC2, PC3)) +
geom_point(aes(fill=PG),shape=21, color="grey") + pp_sra +
guides(color=guide_legend(title="PG")) +
scale_fill_manual(values = colors_CCP)
PCA_CCP_2_3 +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
PCA_CCP_1_4 <- ggplot(prot_pca_x, aes(PC1, PC4)) +
geom_point(aes(fill=PG),shape=21, color="grey") + pp_sra +
guides(color=guide_legend(title="PG")) +
scale_fill_manual(values = colors_CCP)
PCA_CCP_1_4 +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
tsne RNA
genes_no_nas <- multiomics_MAE[["RNAseq_norm"]] %>% is.na() %>% rowSums()
genes_no_nas <- genes_no_nas[ genes_no_nas == 0 ] %>% names()
set.seed(10)
rtsne_out_RNA <- Rtsne( t(assay(multiomics_MAE[genes_no_nas , ,"RNAseq_norm"])), perplexity = 10 )
rtsne_out_RNA_df <- rtsne_out_RNA$Y %>% as.data.frame() %>% as_tibble()
rtsne_out_RNA_df$pat_ID <- colnames(assay(multiomics_MAE[genes_no_nas , ,"RNAseq_norm"]))
rtsne_out_RNA_df <- left_join(rtsne_out_RNA_df,
(wideFormat(multiomics_MAE[, ,c("SNPs","chrom_abber", "health_record_bin") ], colDataCols = c("gender", "treatment_status", "doehner_groups", "PG", "CCP6_RNA")) %>% as_tibble()),
by=c("pat_ID"="primary") )
rtsne_out_RNA_df <- mutate_at(rtsne_out_RNA_df, colnames(rtsne_out_RNA_df %>% dplyr::select(SNPs_ATM:health_record_bin_treated)), as.logical)
rtsne_out_RNA_df <- rtsne_out_RNA_df %>% replace(is.na(.), "unknown")## Warning in `[<-.factor`(`*tmp*`, thisvar, value = "unknown"): invalid factor
## level, NA generatedggplot(rtsne_out_RNA_df, aes(V1, V2)) +
geom_point(aes(color=doehner_groups), size=3) +
scale_color_manual(values = c("#377eb8","#e41a1c", "#984ea3", "#4daf4a", "#ff7f00", "grey"))+
pp_sra +
theme(aspect.ratio=1, plot.title = element_text(size = 30))
tsne_CCP_R_plot <- ggplot(rtsne_out_RNA_df, aes(V1, V2)) +
geom_point(aes(fill=PG), color="grey", shape=21) +
scale_fill_manual(values = colors_CCP)+
pp_sra +
guides(color=guide_legend(title="PG"))
tsne_CCP_R_plot + theme(aspect.ratio=1, plot.title = element_text(size = 30))
tsne_CCPRNA_R_plot <- ggplot(rtsne_out_RNA_df, aes(V1, V2)) +
geom_point(aes(fill=CCP6_RNA), color="grey", shape=21) +
scale_color_hue()+
pp_sra +
guides(color=guide_legend(title="CC_RNA"))
tsne_CCPRNA_R_plot+ theme(aspect.ratio=1, plot.title = element_text(size = 30))
sapply(colnames(rtsne_out_RNA_df)[c(4,5,9:30)], function(var){
print(ggplot(rtsne_out_RNA_df, aes(V1, V2)) +
geom_point(aes_string(color=var), size=3) +
scale_color_manual(values = c("#92c5de", "#f4a582", "grey")) +
ggtitle(gsub("SNPs_", "", gsub("chrom_abber_", "", gsub( "health_record_bin_", "", var) ))) +
pp_sra_noguides +
theme(aspect.ratio=1, plot.title = element_text(size = 30)) )
})
## gender treatment_status SNPs_ATM SNPs_BIRC3 SNPs_EGR2 SNPs_NOTCH1
## data List,30 List,30 List,30 List,30 List,30 List,30
## layers List,1 List,1 List,1 List,1 List,1 List,1
## scales ? ? ? ? ? ?
## mapping List,2 List,2 List,2 List,2 List,2 List,2
## theme List,66 List,66 List,66 List,66 List,66 List,66
## coordinates ? ? ? ? ? ?
## facet ? ? ? ? ? ?
## plot_env ? ? ? ? ? ?
## labels List,4 List,4 List,4 List,4 List,4 List,4
## guides List,4 List,4 List,4 List,4 List,4 List,4
## SNPs_POT1 SNPs_SF3B1 SNPs_TP53 SNPs_XPO1 chrom_abber_del11q
## data List,30 List,30 List,30 List,30 List,30
## layers List,1 List,1 List,1 List,1 List,1
## scales ? ? ? ? ?
## mapping List,2 List,2 List,2 List,2 List,2
## theme List,66 List,66 List,66 List,66 List,66
## coordinates ? ? ? ? ?
## facet ? ? ? ? ?
## plot_env ? ? ? ? ?
## labels List,4 List,4 List,4 List,4 List,4
## guides List,4 List,4 List,4 List,4 List,4
## chrom_abber_del13q14 chrom_abber_del17p13 chrom_abber_del5_IgH
## data List,30 List,30 List,30
## layers List,1 List,1 List,1
## scales ? ? ?
## mapping List,2 List,2 List,2
## theme List,66 List,66 List,66
## coordinates ? ? ?
## facet ? ? ?
## plot_env ? ? ?
## labels List,4 List,4 List,4
## guides List,4 List,4 List,4
## chrom_abber_delIgH_break chrom_abber_gain14q32 chrom_abber_gain8q24
## data List,30 List,30 List,30
## layers List,1 List,1 List,1
## scales ? ? ?
## mapping List,2 List,2 List,2
## theme List,66 List,66 List,66
## coordinates ? ? ?
## facet ? ? ?
## plot_env ? ? ?
## labels List,4 List,4 List,4
## guides List,4 List,4 List,4
## chrom_abber_trisomy12 health_record_bin_elderly_at_diagnosis
## data List,30 List,30
## layers List,1 List,1
## scales ? ?
## mapping List,2 List,2
## theme List,66 List,66
## coordinates ? ?
## facet ? ?
## plot_env ? ?
## labels List,4 List,4
## guides List,4 List,4
## health_record_bin_elderly_patient health_record_bin_gender_binary
## data List,30 List,30
## layers List,1 List,1
## scales ? ?
## mapping List,2 List,2
## theme List,66 List,66
## coordinates ? ?
## facet ? ?
## plot_env ? ?
## labels List,4 List,4
## guides List,4 List,4
## health_record_bin_IGHV_mutated
## data List,30
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4
## health_record_bin_komplex_abberant_karyotype
## data List,30
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4
## health_record_bin_treated
## data List,30
## layers List,1
## scales ?
## mapping List,2
## theme List,66
## coordinates ?
## facet ?
## plot_env ?
## labels List,4
## guides List,4Session Info
sessionInfo()## R version 3.5.2 (2018-12-20)
## Platform: x86_64-apple-darwin15.6.0 (64-bit)
## Running under: macOS Mojave 10.14.6
##
## Matrix products: default
## BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] grid parallel stats4 stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] viridis_0.5.1 viridisLite_0.3.0
## [3] cowplot_1.0.0 survminer_0.4.6
## [5] ConsensusClusterPlus_1.46.0 RColorBrewer_1.1-2
## [7] gplots_3.0.1.1 PMA_1.1
## [9] MultiAssayExperiment_1.8.3 biomaRt_2.38.0
## [11] biomartr_0.9.0 Rtsne_0.15
## [13] IHW_1.10.1 apeglm_1.4.2
## [15] limma_3.38.3 fdrtool_1.2.15
## [17] vsn_3.50.0 forcats_0.4.0
## [19] stringr_1.4.0 dplyr_0.8.3
## [21] purrr_0.3.3 readr_1.3.1
## [23] tidyr_1.0.0 tibble_2.1.3
## [25] tidyverse_1.3.0 TOSTER_0.3.4
## [27] DESeq2_1.22.2 SummarizedExperiment_1.12.0
## [29] DelayedArray_0.8.0 BiocParallel_1.16.6
## [31] matrixStats_0.55.0 Biobase_2.42.0
## [33] GenomicRanges_1.34.0 GenomeInfoDb_1.18.2
## [35] IRanges_2.16.0 S4Vectors_0.20.1
## [37] BiocGenerics_0.28.0 readxl_1.3.1
## [39] ggrepel_0.8.1 ggpmisc_0.3.3
## [41] ggfortify_0.4.8 ggbeeswarm_0.6.0
## [43] effsize_0.7.6 ggpubr_0.2.4
## [45] magrittr_1.5 lemon_0.4.3
## [47] Hmisc_4.3-0 ggplot2_3.2.1
## [49] Formula_1.2-3 survival_3.1-8
## [51] lattice_0.20-38 Matrix_1.2-18
## [53] progress_1.2.2 reshape2_1.4.3
## [55] pheatmap_1.0.12 openxlsx_4.1.4
## [57] gtools_3.8.1 plyr_1.8.4
##
## loaded via a namespace (and not attached):
## [1] tidyselect_0.2.5 RSQLite_2.1.4 AnnotationDbi_1.44.0
## [4] htmlwidgets_1.5.1 munsell_0.5.0 codetools_0.2-16
## [7] preprocessCore_1.44.0 withr_2.1.2 colorspace_1.4-1
## [10] knitr_1.26 rstudioapi_0.10 ggsignif_0.6.0
## [13] labeling_0.3 slam_0.1-46 bbmle_1.0.20
## [16] GenomeInfoDbData_1.2.0 lpsymphony_1.10.0 KMsurv_0.1-5
## [19] farver_2.0.1 bit64_0.9-7 coda_0.19-3
## [22] vctrs_0.2.0 generics_0.0.2 xfun_0.11
## [25] R6_2.4.1 locfit_1.5-9.1 bitops_1.0-6
## [28] assertthat_0.2.1 scales_1.1.0 nnet_7.3-12
## [31] beeswarm_0.2.3 gtable_0.3.0 affy_1.60.0
## [34] rlang_0.4.2 zeallot_0.1.0 genefilter_1.64.0
## [37] splines_3.5.2 lazyeval_0.2.2 acepack_1.4.1
## [40] impute_1.56.0 broom_0.5.2 checkmate_1.9.4
## [43] BiocManager_1.30.10 yaml_2.2.0 modelr_0.1.5
## [46] backports_1.1.5 tools_3.5.2 affyio_1.52.0
## [49] ellipsis_0.3.0 Rcpp_1.0.3 base64enc_0.1-3
## [52] zlibbioc_1.28.0 RCurl_1.95-4.12 prettyunits_1.0.2
## [55] rpart_4.1-15 zoo_1.8-6 haven_2.2.0
## [58] cluster_2.1.0 fs_1.3.1 data.table_1.12.8
## [61] reprex_0.3.0 hms_0.5.2 evaluate_0.14
## [64] xtable_1.8-4 XML_3.98-1.20 emdbook_1.3.11
## [67] gridExtra_2.3 compiler_3.5.2 KernSmooth_2.23-16
## [70] crayon_1.3.4 htmltools_0.4.0 geneplotter_1.60.0
## [73] lubridate_1.7.4 DBI_1.0.0 dbplyr_1.4.2
## [76] MASS_7.3-51.4 cli_2.0.0 gdata_2.18.0
## [79] pkgconfig_2.0.3 km.ci_0.5-2 numDeriv_2016.8-1.1
## [82] foreign_0.8-72 xml2_1.2.2 annotate_1.60.1
## [85] vipor_0.4.5 XVector_0.22.0 rvest_0.3.5
## [88] digest_0.6.23 Biostrings_2.50.2 rmarkdown_1.18
## [91] cellranger_1.1.0 survMisc_0.5.5 htmlTable_1.13.3
## [94] curl_4.3 lifecycle_0.1.0 nlme_3.1-142
## [97] jsonlite_1.6 fansi_0.4.0 pillar_1.4.2
## [100] httr_1.4.1 glue_1.3.1 zip_2.0.4
## [103] bit_1.1-14 stringi_1.4.3 blob_1.2.0
## [106] latticeExtra_0.6-28 caTools_1.17.1.3 memoise_1.1.0